Ran quick tests with beta firmware on my 1.x card.
Usb mode now works, at least cd mode.
Sb mode also works, tested quickly with world of xeen.
Gus mode works, tested with epic pinball.
As always xcom: ufo defence and panzer general does not work in gus mode.

One more beta release, if this goes well I'll release it as v4.1.1 final. https://github.com/polpo/picogus/releases/tag/v4.1.1-beta.1

In this one: fixes bad clicking in single-cycle DMA audio in SB mode (more than the typical clicking that non-autoinit SB titles can have).

I have a small suggestion for future board revisions of PicoGUS - the wavetable connector could use pin numbering to avoid any possible confusion. One pin is marked but to be honest I only noticed it after staring at it for a good while 😁 Maybe the common triangle marking would be more visible

I had little bit time to test my hw v1 PicoGUS yesterday with v4.1.1-beta.1. Focused on DOS demos. Happy to report no issues on VIA Apollo Pro 133T platform. It run just fine even with CPU L1 cache disabled. And yes second reality had stereo sound with SB16 emulation.
Anyways run into issue with Duke3D. In SB16 mode it insist that high DMA must be between 5-7. Hm...
Another observation - Dark Forces imuse setup is reporting SB16 DSP ver 4.05 so PicoGUS emulates very early SB16 version

vutt wrote on 2026-04-23, 15:58:

Anyways run into issue with Duke3D. In SB16 mode it insist that high DMA must be between 5-7. Hm...

Just set it to 5 and than edit the config file and set the high dma to the same value as the low dma. From the PicoGUS wiki:

Because PicoGUS is an 8-bit card, it does not support high DMA for Sound Blaster 16 support. This is actually a mostly-supported configuration for SB16: several versions could be jumpered to disable high DMA, and the Creative Vibra 16X/16XV does not support high DMA at all. Almost all SB16-supporting software is fine working without high DMA, or configuring the same DMA channel for both low and high DMA.

NeoG_ wrote on 2026-04-16, 22:36:

eliot_new wrote on 2026-04-16, 21:24:

Hello, I have plugged in the PicoGUS into an Acorp ALi5 Mainboard. FW 3.2.0. When I load pgsunit with = SB parameter, I hear sound in Doom1, Doom2 normal sound blaster A220, I7, D3 as jumpered on the PicoGUS. But as soon as I select = gus in my boot menu, I don't hear any sound in Doom1, Doom2. This is astonishing, as the same PicoGUS is working with the same MSDOS 6.22 bootmenu on a Asus ALi5 Mainboard without any problems. Tomorrow I will post here my config.sys and autoexec.bat. Maybe some mainboards really have problems, or is it the ISA slot not accepting the PicoGUS?

I am using a PicoGUS in an ACorp 5ALI61 without any issues, so there is going to be some kind of configuration issue somwhere

Ok, here is the solution, it took me almost 1 day to figure it out.

History behind: I bought 2x PicoGUS 2.0 and 3x AWE64 value for cheap.
One pair of PicoGUS 2.0 and AWE64 value worked perfectly 3 months ago along with the onboard ESS1938S Chip on my P5A Rev. 1.4.
Then I bought 2 additional AWE and 1 extra PicoGUS as well as an Acorp ALi5 and a Shuttle HOT-637A i440LX.
After that I began to rotate the soundcards without thinking about the consequences.
I always used the same sd2ide adapter with the same SanDisc 32GB sdcard and MSDOS 6.22 along with Phils Boot Menue and some extra entries into Autoexec.bat for SOLO-1 XOR AWE64 XOR PicoGUS.
I knew that CTCM is saving values into CTPNP.CFG and thought I can easily rotate every sound card to every mainboard.
Finally CTCM began to freeze as well as the PicoGUS software (ultrainit or how was the name?)
I observed the following:
1 of 3 AWE64 worked on Acorp but only in ISA slot number 2, the PicoGUS only in slot number 1, none of them in ISA slot 3 and they always wanted irq 5, dma 1.
All 3 AWE64 and all 2 PicoGUS worked flawlessly on the Shuttle iLX440
None of the 3 nor of the 2 worked on the P5A.

1 hour ago I got a genius idea :
First: I deleted Serial No from CTPNP.cfg
Second: I loaded the defaults into the Bios of P5A
Third: I rebootet into MSDOS 6.22 and suddenly CTCM assigned A220, i10, D3, H7, P330, T6
Fourth: I figured out, that CTMC saved those values into C:\SB16 rather than on C: where I first searched and found an old ctpnp.cfg with i7,D1, H7

The lessons to be learned:

1. i440LX and maybe i440BX, i815EP boards are forgiving and don't mind if you rotate sound cards
2. ALi5 mainboards are like Elephants